​
KlavisToolkit

class KlavisToolkit(BaseToolkit):

A class representing a toolkit for interacting with Klavis API.

This class provides methods for interacting with Klavis MCP server instances, retrieving server information, managing tools, and handling authentication.

Attributes: api_key (str): The API key for authenticating with Klavis API. base_url (str): The base URL for Klavis API endpoints. timeout (Optional[float]): The timeout value for API requests in seconds. If None, no timeout is applied. (default: :obj:None)

​
init

def __init__(self, timeout: Optional[float] = None):

Initialize the KlavisToolkit with API client. The API key is retrieved from environment variables.

​
_request

def _request(
    self,
    method: str,
    endpoint: str,
    payload: Optional[Dict[str, Any]] = None,
    additional_headers: Optional[Dict[str, str]] = None
):

Make an HTTP request to the Klavis API.

Parameters:

  • method (str): HTTP method (e.g., ‘GET’, ‘POST’, ‘DELETE’).
  • endpoint (str): API endpoint path.
  • payload (Optional[Dict[str, Any]]): JSON payload for POST requests.
  • additional_headers (Optional[Dict[str, str]]): Additional headers to include in the request.

Returns:

Dict[str, Any]: The JSON response from the API or an error dict.

​
create_server_instance

def create_server_instance(
    self,
    server_name: str,
    user_id: str,
    platform_name: str
):

Create a Server-Sent Events (SSE) URL for a specified MCP server.

Parameters:

  • server_name (str): The name of the target MCP server.
  • user_id (str): The ID for the user requesting the server URL.
  • platform_name (str): The name of the platform associated with the user.

Returns:

Dict[str, Any]: Response containing the server instance details.

​
get_server_instance

def get_server_instance(self, instance_id: str):

Get details of a specific server connection instance.

Parameters:

  • instance_id (str): The ID of the connection instance whose status is being checked.

Returns:

Dict[str, Any]: Details about the server instance.

​
delete_auth_data

def delete_auth_data(self, instance_id: str):

Delete authentication metadata for a specific server connection instance.

Parameters:

  • instance_id (str): The ID of the connection instance to delete auth for.

Returns:

Dict[str, Any]: Status response for the operation.

​
delete_server_instance

def delete_server_instance(self, instance_id: str):

Completely removes a server connection instance.

Parameters:

  • instance_id (str): The ID of the connection instance to delete.

Returns:

Dict[str, Any]: Status response for the operation.

​
get_all_servers

def get_all_servers(self):

Returns:

Dict[str, Any]: Information about all available MCP servers.

​
set_auth_token

def set_auth_token(self, instance_id: str, auth_token: str):

Sets an authentication token for a specific instance.

Parameters:

  • instance_id (str): The ID for the connection instance.
  • auth_token (str): The authentication token to save.

Returns:

Dict[str, Any]: Status response for the operation.

​
list_tools

def list_tools(self, server_url: str):

Lists all tools available for a specific remote MCP server.

Parameters:

  • server_url (str): The full URL for connecting to the MCP server via Server-Sent Events (SSE).

Returns:

Dict[str, Any]: Response containing the list of tools or an error.

​
call_tool

def call_tool(
    self,
    server_url: str,
    tool_name: str,
    tool_args: Optional[Dict[str, Any]] = None
):

Calls a remote MCP server tool directly using the provided server URL.

Parameters:

  • server_url (str): The full URL for connecting to the MCP server via Server-Sent Events (SSE).
  • tool_name (str): The name of the tool to call.
  • tool_args (Optional[Dict[str, Any]]): The input parameters for the tool. Defaults to None, which might be treated as empty args by the server. (default: :obj:None)

Returns:

Dict[str, Any]: Response containing the result of the tool call or an error.

​
get_tools

def get_tools(self):

Returns:

List[FunctionTool]: A list of FunctionTool objects representing the functions in the toolkit.